Generalized metrics for the analysis of E-learning personalization strategies

For personalizing E-learning, several different strategies and characteristics can be used and considered by teachers and course authors/designers. In order to make appropriate decisions on how to best implement personalized E-learning, this paper focuses on the question: How to foresee personalization strategies that are appropriate for particular courses? To answer this question, we present an approach for recommending personalization strategies based on the learning objects included in the course as well as on how well they support particular combinations of learners’ characteristics. In particular, the paper presents generalized metrics which support teachers for analyzing and comparing personalization strategies, as well as deciding which one should be applied for personalizing each course. The approach was validated through experiments in order to test its feasibility and success when applied to a large number of learning objects and learners’ characteristics.     

Enhancing e-learning systems with personalized recommendation based on collaborative tagging techniques

Personalization of the e-learning systems according to the learner’s needs and knowledge level presents the key element in a learning process. E-learning systems with personalized recommendations should adapt the learning experience according to the goals of the individual learner. Aiming to facilitate personalization of a learning content, various kinds of techniques can be applied. Collaborative and social tagging techniques could be useful for enhancing recommendation of learning resources. In this paper, we analyze the suitability of different techniques for applying tag-based recommendations in e-learning environments. The most appropriate model ranking, based on tensor factorization technique, has been modified to gain the most efficient recommendation results. We propose reducing tag space with clustering technique based on learning style model, in order to improve execution time and decrease memory requirements, while preserving the quality of the recommendations. Such reduced model for providing tag-based recommendations has been used and evaluated in a programming tutoring system.     

Dynamic computerized testlet-based test generation system by discrete PSO with partial course ontology

For the majority of E-learning systems their important role is the electronic teachers who can assist students to learn and review courses enrolled. Unfortunately, an E-learning system does not resemble a teacher who can evaluate the current students’ learning status by the immediate class observations and adjust his/her policy accordingly. Hence, to be able to dynamically compose tests to examine students’ learning status and clear their confusions becomes an important issue. This paper argues that how to formalize test requirements dynamically and how to select suitable items effectively are two main challenges in the current researches of computerized test generation. Hence, this article proposed a novel approach that can compose suitable tests to easily and efficiently evaluate the learning statuses of students. The proposed approach obeys the Discrete Particle Swarm Optimization (DPSO) scheme and uses partial course ontology to generate tests according to the studied materials. Besides, it also harnesses the notion of testlets to construct testlet-based tests for students. In the experiments, the results show that the proposed approach can effectively formalize the tested concepts and their weights, and also generate the well-scrutinized test in accordance with the test requirements in the acceptable time.     

Optimization-based iterative learning for precise quadrocopter trajectory tracking

Current control systems regulate the behavior of dynamic systems by reacting to noise and unexpected disturbances as they occur. To improve the performance of such control systems, experience from iterative executions can be used to anticipate recurring disturbances and proactively compensate for them. This paper presents an algorithm that exploits data from previous repetitions in order to learn to precisely follow a predefined trajectory. We adapt the feed-forward input signal to the system with the goal of achieving high tracking performance—even under the presence of model errors and other recurring disturbances. The approach is based on a dynamics model that captures the essential features of the system and that explicitly takes system input and state constraints into account. We combine traditional optimal filtering methods with state-of-the-art optimization techniques in order to obtain an effective and computationally efficient learning strategy that updates the feed-forward input signal according to a customizable learning objective. It is possible to define a termination condition that stops an execution early if the deviation from the nominal trajectory exceeds a given bound. This allows for a safe learning that gradually extends the time horizon of the trajectory. We developed a framework for generating arbitrary flight trajectories and for applying the algorithm to highly maneuverable autonomous quadrotor vehicles in the ETH Flying Machine Arena testbed. Experimental results are discussed for selected trajectories and different learning algorithm parameters.     

The foundations of a theory-aware authoring tool for CSCL design

One of the most useful ways to enhance collaboration is to create scenarios where learners are able to interact more effectively. Nevertheless, the design of pedagogically sound and well-thought-out collaborative learning scenarios is a complex issue. This is due to the context of group learning where the synergy among learners’ interactions affects learning processes and, hence, the learning outcome. Although many advances have been made to support the designing of collaborative learning scenarios through technology, a more systematic approach is lacking. With the limitations of the current designing methods and tools, it is difficult to develop intelligent authoring systems that can guide users in order to produce more effective collaboration. One of the main difficulties with creating a more consistent (computer-understandable) approach to designing collaboration is the necessity of proposing better ways to formalize the group learning processes. In this paper, we present an innovative approach that uses ontologies and concepts from learning theories to create a framework that represents collaborative learning and its processes. Ontologies provide the necessary formalization to represent collaboration, while learning theories provide the concepts to justify and support the development of effective learning scenarios. Such an approach contributes to establish the foundations for the development of the next generation of intelligent authoring systems referred to as theory-aware systems. To verify the viability and usefulness of our proposed ontological framework in the context of systematic design, the development and use of an intelligent authoring tool for CSCL design is presented. This system is able to reason on ontologies to give suggestions that help users to create theory-compliant collaborative learning scenarios. We carried out several experiments with teachers in a geometry drawing course and the results indicate that the system helps teachers to create and interchange their scenarios more easily and facilitates the selection of important pedagogical strategies that influence positively the designing and effectiveness of group activities.     

Personalized Learning Course Planner with E-learning DSS using user profile

Various methods of E-learning systems, based on information and communications, and geared towards improving learning effectiveness and students’ attention span, have been studied. However, most E-learning systems force students to follow the learning course or content established by a teacher. These methods are convenient, but they limit the effectiveness of E-learning.To overcome this limitation and increase effective learning, new techniques that reflect alternative learning styles, such as adaptive learning and personalized learning, have been studied. In this study, we proposed a Personalized Learning Course Planner (PLCP) that allows students to easily select the learning course they desire. User profile data was collected from the students’ initial priorities about learning contents as well as the test scores after their study. E-Learning Decision Support System (EL-DSS) in PLCP suggests an appropriate learning course organization, according to calculated results based on the user profile data.To verify the effectiveness of the proposed system, we implemented an English learning system consisting of PLCP. We conducted an experiment with 30 university students and evaluated students’ satisfaction by questionnaire analysis. The results indicate that the proposed system improved learning effectiveness and student satisfaction. Further investigation of the participants indicated that suggesting a learning course suitable for students’ previous test scores and priorities encouraged students to concentrate on the lesson.     

Architecting centralized coordination of soccer robots based on principle solution

Coordination strategy is a relevant topic in multi-robot systems, and robot soccer offers a suitable domain to conduct research in multi-robot coordination. Team strategy collects and uses environmental information to derive optimal team reactions, through cooperation among individual soccer robots. This paper presents a diagrammatic approach to architecting the coordination strategy of robot soccer teams by means of a principle solution. The proposed model focuses on robot soccer leagues that possess a central decision-making system, involving the dynamic selection of the roles and behaviors of the robot soccer players. The work sets out from the conceptual design phase, facilitating cross-domain development efforts, where different layers must be interconnected and coordinated to perform multiple tasks. The principle solution allows for intuitive design and the modeling of team strategies in a highly complex robot soccer environment with changing game conditions. Furthermore, such an approach enables systematic realization of collaborative behaviors among the teammates.     

Re-purposing existing generic games and simulations for e-learning

There is a growing interest among teachers in using games as a part of their lesson plans. A standardised, interoperable approach to the sharing of such game-based lesson plans would allow teachers and educational technologists to compare and contrast Digital Game Based Learning scenarios, allowing best practices and lessons learned to emerge. Although games can be used as ‘add-ons’ in educational contexts, greater benefits can be attained by integrating games more fully into the educational process, i.e. by repurposing existing games to target the specific learning objectives. In this article we analyse this problem. We developed two possible solutions based on the integration and the interaction of games and learning scenarios. The first solution is based on ‘pedagogical wrappers’, where games are linked to e-learning flows but without interaction and communication. The second solution sees a tighter integration which supports ongoing interaction and communication between game and e-learning flow. We applied both solutions to a generic game. This game was firstly programmed in Action Script and later re-used for learning purposes and represented in IMS Learning Design. We analysed the pros and cons of each solution and identify research topics for further research.     

EDUCA: A web 2.0 authoring tool for developing adaptive and intelligent tutoring systems using a Kohonen network

This paper presents a Web 2.0 Learning Environment, for a systematic creation of adaptive and intelligent tutoring systems. Authoring contents is made by a community of users including teachers and students. The tutoring systems adapt the contents according to the best learning style using self-organizing maps (SOMs). The SOM was trained for classifying Felder–Silverman learning styles. The most important advantage of these unsupervised neural networks is that they do not require an external teacher for presenting a training set. The approach was implemented under an authoring tool that allows the production of personalized learning material to be used under collaborative and mobile learning environments. The tutoring systems together with the neural network can also be exported to mobile devices. We present different results to the approach working under the authoring tool.     

Design and evaluation of planning and mathematical models for generating learning paths

Education research has emphasized the need to develop instructional design tools to facilitate the generation of learning paths for students. Learning paths are important because they enable the personalization and optimization of the learning process. In this work, we present a flexible conceptual framework that allows the representation of curricula information as Artificial Intelligence Planning and Mathematical Programming models to facilitate the generation of learning paths by domain independent algorithms. The resulting models consider a rich set of properties from the education domain, like hierarchical learning structures, enabling conditions, temporal actions, mandatory activities, quality accumulation functions, and metric information. We show that the proposed mathematical models return optimal solutions very efficiently if we relax the total ordering constraints of learning paths. These relaxations allow evaluating greedy planning algorithms to identify the properties from the models that increase the complexity of solution synthesis. We expect that the results of this research can be helpful to education researchers and computer scientists in the quest of scalable systems that capture more flexible standards to model learning and compute more informed learning paths for students.     

本体理论及其在E-Learning中的应用

E-learning是信息时代的产物,是学习形式、学习手段和学习方法的革命.但是,E-learning面临一个重大挑战,即如何表示教学内容的丰富语义.根据现代认知心理学理论,从教育教学规律出发,提出一种教学领域本体的构建方法.借助该方法,设计和建立了C类程序设计课程领域本体.实践证明它能满足用户自适应学习的需求,降低教师的教学难度和学生的学习难度.构建的领域本体对“C程序设计”和“C++程序设计”这两门课程知识组织结构形成了共同理解与认识,为进一步建立支持个性化语义检索的课程资源网络系统奠定了基础.     

User Profile Modeling in the context of web-based learning management systems

Over the past two decades, great research efforts have been made towards the personalization of e-learning platforms. This feature increases remarkably the quality of the provided learning services, since the users’ special needs and capabilities are respected. The idea of predicting the users’ preferences and adapting the e-learning platform accordingly is the focal point of this paper. In particular, this paper starts with the main requirements of an advanced e-learning system, explains the way a user navigates in such a system, presents the architecture of a corresponding e-learning system and describes its main components. Research is focused on the User Model component, its role in the e-learning system and the parameters that comprise it. In this context, Bayesian Networks are used as a tool for the encoding, learning and reasoning of probabilistic relationships, with the aim to effectively predict user preferences. In support of this vision, four different scenarios are presented, in order to test the way Bayesian Networks apply in the e-learning field.     

Creating a performance-oriented e-learning environment: A design science approach

E-learning is now being used by many organizations as an approach for enhancing the skills of knowledge workers. However, most applications have performed poorly in motivating employee learning, being perceived as less effective due to a lack of alignment of learning with work performance. To help solve this problem, we developed a performance-oriented approach using design science research methods. It uses performance measurement to clarify organizational goals and individual learning needs and links them to e-learning applications. The key concept lies in a Key Performance Indicator model, where organizational mission and vision are translated into a set of targets that drive learning towards a goal of improving work performance. We explored the mechanisms needed to utilize our approach and examined the necessary conceptual framework and implementation details. To demonstrate the effectiveness of the approach, a prototype workplace e-learning system was developed and used to evaluate the effectiveness of our approach.     

A system for knowledge discovery in e-learning environments within the European Higher Education Area – Application to student data from Open University of Madrid,UDIMA

In today's open and dynamic learning environment, a significant percentage of students have a preference for flexible learning systems whereby they can reconcile their academic pursuits with their job responsibilities and family obligations.Non face-to-face educational models, like e-learning (electronic learning), evolved in order to offer such flexibility. E-learning systems have major strengths but also pose major challenges to the educational community.One such challenge is the large spatial and temporal gap between the teacher and student, which is an obstacle to student follow-up by teachers. The information generated by virtual learning systems sometimes overwhelms instructors who are unable to process the data without the support of special-purpose techniques and tools that are useful for analysing large dataflows.     

开启联邦学习的个性化时代：隐私、安全与效率的平衡探索

人工智能为公共和国防安全的发展和应用提供了巨大的机遇,然而国防安全数据包含了敏感的军事、情报和战略信息,一旦泄露或被滥用,可能对国家安全造成严重威胁,如何确保国防安全数据的隐私保护成为一项重要挑战。个性化联邦学习是近年来发展起来的一种新型的机器学习方法,它旨在通过将分布式的数据在本地进行训练和更新,从而实现在保护数据隐私的前提下提高本地模型的准确性和鲁棒性。与传统的中心化机器学习方法不同,个性化联邦学习允许不同数据拥有者之间共享模型的信息,而不是数据本身。这种方法已经在医疗、金融、物联网等领域得到了广泛的应用。本文从全局模型个性化和本地模型个性化两个方面分别介绍了个性化联邦学习的基本原理以及研究现状,总结了各个方法的优缺点,并讨论了现有方法的评价指标和常用数据集,最后展望了它在未来的发展前景。     

Trading strategy design in financial investment through a turning points prediction scheme

Turning points prediction has long been a tough task in the field of time series analysis due to its strong nonlinearity, and thus has attracted many research efforts. In this study, the turning points prediction (TPP) framework is presented and further employed to develop a novel trading strategy designing approach to financial investment. The TPP framework is a machine learning-based solution incorporating chaotic dynamic analysis and neural network modeling. It works on the ground of a nonlinear mapping deduced in financial time series through chaotic analysis. An event characterization method is created in TTP framework to characterize trend patterns in ongoing financial time series. The main contributions of this paper are (1) it presents an ensemble learning based TPP framework, within which the nonlinear mapping is approximated by the ensemble artificial neural network (EANN) model with a new parameters learning algorithm; (2) a genetic algorithm (GA) based threshold optimization procedure is described with a newly defined performance measure, named TpMSE, which is used as a cost function; and (3) a trading strategy designing approach is proposed based on the TPP framework. The proposed approach was applied to the two real-world financial time series, i.e., an individual stock quote time series and the Dow Jones Industrial Average (DJIA) index time series. Experimental results show that the proposed approach can help investors make profitable decisions.     

本体在E-learning系统中的应用研究

探讨了E-learning及本体，提出了基于本体的E-learning系统层次结构模型，并重点研究了本体在其中的应用：用于描述学习材料语义的内容本体，用于定义学习材料上下文的上下文本体以及用于在学习课程中组织学习材料的结构本体。     

An ACO-based personalized learning technique in support of people with acquired brain injury

The ever-increasing cases of acquired brain injury (ABI), especially among young people, have prompted a rapid progress in research involving neurological disorders. One important path is the concept of relearning, which attempts to help people regain basic motor and cognitive skills lost due to illness or accident. The goals of relearning are twofold. First, there must exist a way to properly assess the necessities of an affected person, leading to a diagnosis, followed by a recommendation regarding the exercises, tests and tasks to perform; and second, there must be a way to confirm the results obtained from these recommendations in order to fine-tune and personalize the relearning process.This presents a challenge, as there is a deeply-rooted duality between the personalized and the generalized approach. In this work we propose a personalization algorithm based on the ant colony optimization (ACO), which is a bio-inspired meta-heuristic. As we show, the stochastic nature of ants has certain similarities to the human learning process.We combine the adaptive and exploratory capabilities of ACO systems to respond to rapidly changing environments and the ubiquitous human factor. Finally, we test the proposed solution extensively in various scenarios, achieving high quality results.     

混合式学习支撑系统的设计与实现

针对目前高等学校教育的特点,设计并开发一个混合式学习支撑系统,给出其总体架构、概念模型及数据库的设计。该系统结合课堂教学和在线学习优势,既能支持教师教学,又能支持学生自主学习能力和创新能力的培养,可根据学生兴趣和教学资源的形态,自动选择最适合其使用该教学资源的教学策略,提高学习效率。     

对于E-Learning的特性分析与应用探究

随着计算机网络的普及,E-Learning正在被越来越多的行业领域和人们所接受。而我国高校本身就是一个巨大的教学资源基地,在高校开展E-Leamirtg教学资源环境下进行更有效的学习已经成为一种必然的趋势,本文通过对E-learning含义的阐述、E-learning的需求分析以及特性分析来阐述这一论断,从而使E-learning能够更多样化的服务于高校环境下教师与学生。